STRUCTURED OBSERVATION
Structured Observation is potentially one of the most useful field methods in drug use studies. This chapter is devoted to a brief overview of the method and a detailed description of the main steps involved in conducting an observational drug use study.
5.10:Overview
5.11: What is Structured Observation
Observation is a technique that involves directly observing behaviour with the purpose of describing it. To observe means to examine an object, or an individual, or group of people, or an event with all of the senses. Recording of observations may take many forms, from simple and casual to exact and sophisticated. For example, an observer may observe an event and then complete a checklist on whether or not key behaviours occurred. Or the observer may write notes on everything that happens in his or her presence. More sophisticated recording may involve audio-visual devices.
The technique can be classified into participant and non-participant observations. Participant observation takes place when an observer participates with the people and in the events he or she is observing. Non-participant observation occurs when an observer observes events without interacting with the person(s) being observed.
Non-participant observation may further be classified as structured or unstructured. The aim of unstructured observation is to observe and record behaviour in a holistic way without the use of a pre-determined guide. Structured observation, on the other hand, refers to a technique in which an observer observes events using a guide that has been planned in advance. The focus of this chapter is on structured observation.
Events in structured observation are recorded according to an observation guide. The observer is not involved in the activities being observed, but records them as inconspicuously as possible. However, it must be pointed out that the presence of even a "neutral" and non-interacting observer may influence the behaviors of the person(s) or events being observed.
5-1
Although structured observation may seem simple, obtaining useful data requires reliable observers, an informative observational guide, and the cooperation of those being observed. This presents the issue of bias in the observation process in both the observer and the person being observed. The observer's bias includes his or her subjective judgment regarding events being observed. For example, if the observer has a positive attitude towards the person being observed, he or she may record positive observations and ignore negative behaviours. To overcome this bias, an observer must be trained to be neutral and non-judgmental towards persons being observed.
From the perspective of the person being observed, bias occurs when the person being observed alters his or her usual performance of an activity in order to impress the observer. In a diarrhoea study in one country for example, when prescribers were informed that the observation concerned diarrhea, they examined infants who had diarrhea longer than they did other patients. The examination took about 5-7 minutes for diarrhea cases, whereas the examination for other patients took only 1-2 minutes.
Surrogate Patient Technique
To overcome certain problems of bias, some studies employ the use of surrogate patients. In this technique a client suitably dressed and prepared presents at a health facility or dispensing/drug sale outlet with a complaint requiring treatment. The complaint may concern the same individual or the treatment may be for another person, such as a child. When combined with direct interviews, the method affords an investigator the opportunity to contrast what providers say they do with what they actually do. The reasons for the discrepancies can then be ascertained using in-depth interviews or focus groups.
5.12: Use of Structured Observation in Drug Use Studies
In drug use studies structured observation can be used independently or as supplement to other methods. As an independent method, it can be used to observe situations in health facilities. This will involve part or all the sequence of events commencing when a patient comes to register until he or she leaves the health center. In general, the goal for using the observational method in drug use studies involves one or more of the following:
!To generate research hypotheses in pilot studies where very little is known about the problem.
!To collect information not available in any other way, such as communication patterns during a provider-patient encounter.
!To supplement other data as part of a multiple assessment approach where observation may aid in the interpretation of data.
As a supplementary method, structured observation can also be made during focus group discussions (FGDs), in-depth interviews, or interviews using a structured questionnaire. It can also be used to supplement quantitative methods of data collection, such as prescribing surveys.
5-1
5.20:Key Steps in Conducting Structured Observations
The steps listed in Table 5.1. can be followed in applying the structured observational method to study drug use problems. The list is followed by a more detailed discussion of each step.
TABLE 5.1:Key Steps in Conducting Structured Observations
Step 1.Decide if a resource person is needed.Step 2.Determine what is to be observed.
Step 3.Choose the observers.
Step 4.Develop observation guides.
Step 5.Select the setting for the observations.
Step 6.Train observers and pre test observations.
Step 7.Conduct the observations.
Step 8.Analyze and interpret the observational findings.
5.21. STEP 1: Decide If a Resource Person Is Needed
Use of structured observations in a drug use study requires certain skills which those involved in the clinical aspects of health may lack. One of the first decisions to make is whether the assistance of a resource person is needed. The following are some points to consider.
a.Assessing the Need for a Resource Person
The assistance of a resource person is very helpful, but whether you need one or not will depend on a number of factors. These include size of the study, the resources available for it, and the experience of the investigator and others on the study team.
If the study is small, you may rely on the services of someone who can consult for brief periods when you need his/her services. However, if it is a large study you may need to look for someone who can devote sufficient time to the study. The use of a resource person can be very important to the success of the study, particularly if you have no experience with observational methods.
5-1
b.Who to consider as a Resource Person
The role of a resource person is to assist in planning the observation protocol and preparing for subsequent stages of the study. The first choice for a resource person would be a social scientist who has the knowledge, skills, and actual field experience in using structured observational methods. Where it is difficult to find a social scientists to assist, others whose jobs are behaviourally oriented and who have actually done some work using observations such as colleagues in public health or community development officers, can be equally useful. In any case, the resource person needs to have knowledge about the health care system and the context of drug use in the locality.
5.22. STEP 2: Determine What Is to Be Observed
The beginning of preparing an observation protocol is to determine what behaviours or encounters will be observed on the field. The following describes some guidelines.
a.Selection Criteria
As with all applied qualitative methods, the purpose of the study will determine who and what is to be observed. Since structured observation is often used to validate data obtained from other methods, appropriate sources of information must be identified. For example, suppose the purpose of an observation encounter is to verify a claim by providers that they give antibiotics to children who present with ARI because their accompanying parents ask for them. Since you cannot observe all patients in the targeted population, you will need to develop a method to select encounters to represent this group.
Most clinic-based observational studies use a convenience sample of patients who happen to be available at the time of data collection. Thus, all patients who report with cough on a particular day at a certain health center might be observed. A random sample of patients with cough in the community during a given period would be impractical to obtain.
b.Sources of Information
An investigator using structured observations must decide beforehand what the target of attention of the study will be. Pertinent issues to consider before designing the observation protocol include:
!Is the emphasis to be on drug prescribing habits, adequate patient care, patient behaviour, the health system context, or all of these?
!Will the focus of attention be the health provider, the patient, or the prescribing encounter?
!Will the observation take place in hospitals, health centers, private drug outlets, or some other place?
5-1
!Will the observation cover the interaction of prescriber and patient, or the whole situation at the health facility or drug outlet?
!What are the possible opportunities in the observing environment for the observer to blend in and become less obtrusive?
!What is the most appropriate recording system for this situation?
The final decisions must be based upon the aims and goals of the study and the practical aspects of collecting the data. Remember to narrow attention to data that are useful in designing an intervention, and not other interesting but possibly unnecessary information.
5.23.STEP 3: Choose the Observers
In preparing for data collection, you need to select the most appropriate field staff to carry out the observations. The following are essential prerequisites to consider when recruiting field staff.
a.Qualities to Consider When Selecting Observers
The following observer qualities are helpful in ensuring accurate data:
!Familiarity with the cultural background of the people being observed and ability to understand their language is a key qualification.
!Familiarity with pharmaceutical and general medical terms is a benefit, and may be an absolute prerequisite for certain studies..
!Knowledge about social research techniques is an advantage.
!A secondary or high school education may be sufficient if the medical aspects of the behaviour being observed are not complex..
!Observers must be able to sit quietly and observe without interfering.
Ideally, structured observation would use two observers to ensure reliable information.
5-1
However, it may not always be possible to have two observers for every situation. Besides, it may be more difficult for two observers to stay "unnoticed" than one person. With adequate training, a single observer should be able to produce high-quality and consistent data.
b.Observer's Role in Structured Observation
In structured observations, the observer watches events unfold without affecting them and records these events in appropriate categories (on an already prepared checklist). The observer is like a video recorder. He or she should record as objectively as possible all relevant events, situations, and behaviours.
To minimize bias, the observer tries to be as unobtrusive as possible so that subjects feel at ease and comfortable in carrying out their daily activities. The best way is to "sit quietly as a fly on the wall in the corner" so that the subject forgets that the observer is in the same room. The observer must maintain interest in the events being observed. Otherwise he or she will carry out the observation mechanically and risk failing to notice some important events.
5.24.STEP 4: Develop Observation Guides
The basis for structured observations is the observation guide. A preliminary list of issues to be observed, and the categories in which to record them, must be prepared in advance based on the objectives of the study. A list of common items that can be observed during clinical encounters at a health facility is included in Table 5.2.
a.Who Prepares the Guide?
Usually the investigator(s) and/or the resource person directing the study is/are the appropriate persons responsible for writing the guide. However, it is advisable and helpful to involve the observers to enable them have input and familiarize themselves with the study. Useful input might also be obtained from people who would have special insight into the interactions being observed (e.g., experienced clinicians, community leaders), or administrators who will be involved in later interventions.
b.Structure and Contents of Guide
Since it is difficult to predict observable behavior in advance, a draft protocol can initially be partially structured. More attention should be paid to specific aspects that appear central to the problem as the study proceeds, and a final observation guide prepared.
The guide should be prepared to suit local situations; the input of the field team is particularly useful. You may also find it useful to begin with the instruments and protocols presented in the annex, and adapt them to the local environment and problem of interest.
5-1
Table 5.2 Items That Can Be Observed During Clinical Encounters in a Health Facility
1.REGISTRATION:!impressions
payment
2.SCREENING:
!does it occur?
!who does it?
!content
3.WAITING:
!what do parents and patients do?
!any health education opportunities?
!how long?
4.CLINICAL CONSULTATION:
!physical environment: private, others present, place for patient to sit, exam. bed,
greeting ritual
!medical history: duration, associated symptoms (fever, chills, chest pain, cough, yellowish sputum in cough), appetite (eating, drinking), previous treatment
!clinical exam: temperature, pulse, breathing (rate, in-drawing of space between the ribs), touching, listening to the chest.
!use of instruments(stethoscope, BP machine, etc.)
!interaction: warm, cold, patient encouraged to speak, type of language (technical vs. non-technical, local vs. non-local)
!explanation of illness: causes, prognosis
!advice: about prevention (environment, hand washing, food storage, clean water, latrines),
!about feeding (fluids, breast feeding), explanation about drugs
!length of time
!how encounter is terminated: prescribing on termination, reassurance, warm, abrupt
!prescriber washes hands after examination.
5.WAIT FOR DRUGS:
!how long?
!what happens during the waiting period?
6.DISPENSING ENCOUNTER:
!physical environment: private, possible to talk with dispenser
!how are drugs dispensed: packaging, sterile technique for injection, labeling
!communication: instructions about how to take, side effects/precautions, prevention and care, patient asks questions
5-1
c.Useful Events to Observe During Clinical Encounters
The most frequent target of observation in drug use studies is the clinical encounter between health provider and patient. The following are key events that can be observed during the clinical encounter:
1.Aspects of Clinical Process
Polypharmacy
The focus of the observation is on how many drugs a provider prescribes for a patient and why. Observation can take place in a health center, or in clinical examination areas during a working day. The observer can sit quietly close by the prescriber and record the drugs being given to the patient and any discussion about reasons for multiple drug therapy.
Non-pharmaceutical therapy
The focus of the observation is on any instances in which a provider consults with a patient without prescribing any drugs, and why. In drug use studies, the observation usually takes place in the consulting room, although observing interactions between patients and nurses or dispensers may also be informative..
Selection of Drugs
The focus of the observation is on the choice of drugs, and reasons for specific choices. Many different aspects of the decision-making process can be covered: adequacy of history taking and examination; discussion of alternatives with patients; patient requests for particular drugs; the use of generic names; stock availability; and so forth.
Two prescribing choices that have received particular attention are use of antibiotics and injections. Which antibiotics are being given to patients, and for what diagnoses? What are the opportunities for laboratory investigation, and are they used appropriately? For injections, in addition to why they are given and whether they are necessary, observation can also focus on adequacy of sterile practice.
Information about use, risks, side effects
The focus of the observation is on the information communicated to the patient about prescribed drugs, on both whether communication takes place at all and whether it is adequate. Issues addressed can include how and when to take the drugs prescribed, possible side effects and what to do if they occur, as well as responses to patient questions or concerns. These interactions can be observed in the consultation room as well as in the dispensing room.
5-1
2.Features of provider-patient interaction
In addition to the more clinical aspects of the encounters between patients and health providers, the observational method can be used to study the nature of the interaction between them. It is often non-clinical aspects of the encounter that determine the overall quality of care from the patient=s perspective. The following aspects of the encounter can be included in the observation guide.
!exchange of greetings between provider and patient;
!verbal expressions of concern by the provider;
!conversation between provider and patient;
!nonverbal expressions such as smile, touch, tone of voice;
!eye contact between provider and patient;
!duration of the encounter;
!language used during the encounter;
!advice given by the provider;
!questions posed by the patient;
!how provider ends the interaction.
5.25. STEP 5: Select the Setting for the Observations
Before beginning the field work, you need to decide the sites, days, times, and other related issues for the observations. The following steps outline the preparations that need to be made.
a.Site Selection and Permission
Some days or weeks before the field work, the investigator should visit the location of the study to learn about the setting and people in order to make a final selection of study sites. When this visit is made, it is possible to obtain permission to carry out the observations from relevant authorities. The meeting can be used to inform local leaders about the purpose of the study, and how it is intended to benefit them, and how many observers will be involved.